(Circulation. 1995;91:2454-2469.)
© 1995 American Heart Association, Inc.
Articles |
Nonstationary Vortexlike Reentrant Activity as a Mechanism of Polymorphic Ventricular Tachycardia in the Isolated Rabbit Heart
From the Department of Pharmacology, SUNY Health Science Center at Syracuse, NY, and the Department of Theoretical Biology (A.P.), University of Utrecht, Utrecht, the Netherlands.
Correspondence to Richard A. Gray, SUNY Health Science Center, Department of Pharmacology, 766 Irving Ave, Syracuse, NY 13210.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background Ventricular tachycardia may result from vortexlike reentrant excitation of the myocardium. Our general hypothesis is that in the structurally normal heart, these arrhythmias are the result of one or two nonstationary three-dimensional electrical scroll waves activating the heart muscle at very high frequencies.
Methods and Results We used a combination of high-resolution video imaging, electrocardiography, and image processing in the isolated rabbit heart, together with mathematical modeling. We characterized the dynamics of changes in transmembrane potential patterns on the epicardial surface of the ventricles using optical mapping. Image processing techniques were used to identify the surface manifestation of the reentrant organizing centers, and the location of these centers was used to determine the movement of the reentrant pathway. We also used numerical simulations incorporating Fitzhugh-Nagumo kinetics and realistic heart geometry to study how stationary and nonstationary scroll waves are manifest on the epicardial surface and in the simulated ECG. We present epicardial surface manifestations (reentrant spiral waves) and ECG patterns of nonstationary reentrant activity that are consistent with those generated by scroll waves established at the right and left ventricles. We identified the organizing centers of the reentrant circuits on the epicardial surface during polymorphic tachycardia, and these centers moved during the episodes. In addition, the arrhythmias that showed the greatest movement of the reentrant centers displayed the largest changes in QRS morphology. The numerical simulations showed that stationary scroll waves give rise to monomorphic ECG signals, but nonstationary meandering scroll waves give rise to undulating ECGs characteristic of torsade de pointes.
Conclusions Polymorphic ventricular tachycardia in the healthy, isolated rabbit heart is the result of either a single or paired ("figure-of-eight") nonstationary scroll waves. The extent of the scroll wave movement corresponds to the degree of polymorphism in the ECG. These results are consistent with our numerical simulations that showed monomorphic ECG patterns of activity for stationary scroll waves but polymorphic patterns for scroll waves that were nonstationary.
Key Words: tachycardia • reentry • arrhythmia • torsade de pointes
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Polymorphic ventricular tachycardia (PVT) is a particularly dangerous unstable arrhythmia that is thought to be a precursor to ventricular fibrillation. It is characterized by a complex ECG pattern with irregular QRS morphology at high rates. The mechanism of PVT, however, remains unclear. As recently reviewed by Davidenko,1 proposed hypotheses include both nonreentrant and reentrant mechanisms. One nonreentrant mechanism that has been proposed is triggered activity brought about by early afterdepolarizations.2 Another mechanism is based on the idea of a single idioventricular focus firing at a constant period but resulting in Wenckebach patterns of activation due to progressively longer ventricular refractoriness.3 Other possibilities that have been suggested are widely spaced arrhythmogenic foci4 and an automatic focus that migrates through the myocardium.5 Reentrant mechanisms that have been proposed for PVTs and torsade de pointes are nonstationary scroll waves (three-dimensional spiral waves) of electrical activity.1 6 7 Experimental results consistent with this proposal have been provided by several studies.8 9 10 Other reentrant circuit variations presented to explain PVT are intermittent conduction block from an exit site of a reentrant pathway11 ; varying exit sites from a reentrant circuit12 13 ; two reentrant circuits, either in close proximity to each other14 15 or far apart5 ; and conduction reversal of a reentrant circuit.16
Spiral wave reentry can be initiated in thin slices of cardiac tissue.17 Spiral waves rotate around a phase singularity (core) where isochrones converge and conduction velocity is slow. The size of this core or "rotor" is determined by the curvature of the wave front and by the refractory period of the medium.18 Recent video imaging experiments in thin slices of sheep epicardial muscle have led to the hypothesis that regardless of the initiating event, vortexlike reentrant activity may in fact underlie a large proportion of both monomorphic and polymorphic ventricular tachycardias.1 7 19 Indeed, depending on spiral core dynamics, monomorphic, undulating, or completely irregular ECG patterns may be observed.1 7 19 Moreover, transitions between such patterns can also occur. For example, drifting spirals giving rise to polymorphic activation can become stationary and result in monomorphic patterns as a result of "anchoring" of the core to a small discontinuity (for example, an artery or a small scar) in the tissue. Such studies have led to the suggestion that the behavior of the spiral center (the rotor) may play a key role in determining the ECG manifestation of the arrhythmia.1 19
Since the heart is in fact three-dimensional, the spiral wave activity observed in very thin sheets of cardiac tissue cut from the surface of the ventricle can only be used as an approximation to the real-life situation. Winfree20 provided the first demonstration of three-dimensional (3-D) spiral waves (scroll waves) in thick layers of the so-called "Belousov-Zhabotinsky" reaction. Simple scroll waves may be constructed by stacking spiral waves. Successive slices of spiral waves may be rotated and slightly shifted, but the line connecting the cores must be continuous. The line connecting the cores is called a filament, and scroll wave filaments come in many shapes21 : they can be linear (straight), L-shaped, U-shaped, or even ring-shaped (the filament forms a closed loop in this case). Scroll wave filaments can also become twisted, in which case exceedingly complex dynamics may occur, depending on the number of twists.22 These different filament shapes are important because the manifestation of the activity on the surface of the 3-D medium will depend on the dynamics of the scroll wave filament.23
There is considerable evidence from surface electrogram recordings that
reentrant activity occurs during cardiac
arrhythmias.8 24 25 26 27 28
However, there is little information
in the electrophysiological literature about whether the myocardium is
able to sustain 3-D scroll waves, which probably reflects the inability
to map excitation across the wall with sufficiently high resolution.
The experiments of Chen et al9 and Frazier et
al,10 however, provide support to the idea that scroll
waves can be induced in the healthy heart. These investigators obtained
transmural recordings of the activation patterns of the right
ventricular outflow tract of the in situ canine heart during and
immediately after the application of cross-field stimulation
(S1-S2) through long electrodes sutured to the
epicardial surface over the recording electrodes. The isochrone maps
obtained from epicardium, midmyocardium, and endocardium demonstrated
vortexlike activity (period of 
90 to 110 milliseconds) throughout
the thickness of the outflow tract wall, suggesting that the
stimulation protocol resulted in the formation of a scroll wave whose
filament was nearly perpendicular to the
surface.10 29
Our studies were directed toward determining whether, in the structurally normal heart, PVT is the result of nonstationary scroll waves. Thus, we used the Langendorff-perfused rabbit heart, a voltage-sensitive dye, and a video camera to map the transmembrane potential changes on the epicardial ventricular wall and to determine whether spiral waves are demonstrable on the surface of the isolated rabbit heart undergoing ventricular tachycardia (VT). Conceivably, such spirals are the two-dimensional (2-D) epicardial surface representation of scroll waves spanning the ventricular wall. In fact, on the basis of the theory of wave propagation in excitable media, we surmise that perhaps monomorphic VT and PVT are both the result of scroll waves and that these scroll waves are not stationary for PVT. To complement these studies, we used a mathematical model based on FitzHugh-Nagumo kinetics and realistic heart geometry30 to study how stationary and nonstationary scroll wave patterns of activity are manifest on the heart surface and in the simulated ECG. Our simulations show that stationary 3-D scroll waves give rise to stationary surface patterns of reentrant activity (spiral waves) and monomorphic ECG signals. On the other hand, nonstationary 3-D scroll waves give rise to nonstationary surface patterns of spiral waves and polymorphic ECG signals. Our experimental data exhibit similar surface patterns and ECGs. The overall results strongly support the idea that in the isolated rabbit heart, PVT results from nonstationary scroll waves of electrical activity.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Experimental Approach
Langendorff-Perfused Rabbit Heart
All experiments were conducted in isolated, Langendorff-perfused rabbit hearts. New Zealand rabbits (
2 kg) were anesthetized with
sodium pentobarbital (35 mg/kg). The heart was rapidly removed through
a thoracotomy and rapidly connected to the Langendorff apparatus; the
coronary arteries were continuously perfused via a cannula in the
aortic root with warm (37° to 39°C) HEPES-Tyrode solution buffered
to a pH of 7.4 under a pressure head of 70 mm Hg. The solution
consisted of the following (mmol/L): NaCl, 148; KCl, 5.4;
CaCl2, 1.8; MgCl2, 1.0;
NaHCO3, 5.8; NaH2PO4,
0.4; glucose, 5.5. The solution was saturated with 100% oxygen, and
albumin (40 mg/L) was added to reduce the possibility of edema.
Subsequently, the heart was immersed in a circular beaker full of warm
HEPES-Tyrode solution, which acted as a volume conductor for recording
the ECG. We used a horizontal lead system that was connected to an
amplifier (Gould, universal amplifier) and bandpass filtered at 0.05 to
1000 Hz. The ECG recordings were displayed on a digital oscilloscope
(Tektronix model 2214) and transferred to a computer (Gateway 386/33)
via a serial connection. At least 20 minutes was allowed for equilibration to ensure that the heart was in sinus rhythm and contracting forcefully and rhythmically. After this, Tyrode solution containing the potentiometric dye di-4-ANEPPS (15 µg/mL) dissolved in DMSO was perfused through the coronaries for 1 to 2 minutes. To stop the contraction of the heart and thus record the fluorescence associated with the transmembrane electrical activity in the absence of mechanical artifacts, we continuously perfused the coronary arteries with HEPES-Tyrode solution containing diacetyl monoxime (DAM, 10 mmol/L). Our recently published experiments have demonstrated that DAM is a reliable electromechanical uncoupler7 17 that has unspecific but relatively minor and reversible effects on the transmembrane currents that control action potential duration.31
Temperature Control
Temperature was constantly monitored by a probe (YSI model 520)
inserted into the right ventricle and connected to a telethermometer.
Our Langendorff perfusion setup was equipped with two
temperature-controlled columns. These columns were used to heat the
Tyrode solution to various temperatures (39° and 32°C). A stopcock
switch allowed us to change the temperature of the perfusing solution
in the coronaries within 10 seconds.
High-Resolution
Optical Mapping
A diagram of the experimental setup is presented in
Fig 1
. The light from a tungsten-halogen lamp was collimated
and made quasimonochromatic by the use of an interference filter (520
nm) together with a heat filter. The light was then reflected 90°
from a dichroic mirror (560 nm) and reflected by another 90° by a
standard mirror so that the light shone on the epicardial surface of
the vertically hanging heart. The emitted fluorescence caused by
transmembrane potential changes of cardiac cells was collected with a
50-mm objective lens with a depth of field of 12 mm. The emitted
light was transmitted through the emission filter (645 nm) and
projected onto a CCD video camera (Cohu). The video images (typically
200x200 pixels) of the epicardial surface were acquired with an
analog-to-digital frame grabber (Epix) in a noninterlace mode with a
speed of 60 frames per second (16.67-millisecond sampling rate). The
heart was initially positioned such that the left anterior descending
coronary artery was facing the light source and the video camera (see
Fig 1). In some experiments, the heart was rotated by hand to
record
from various heart surfaces. The spatial resolution varied with the
magnification but was 0.15 mm. The frame grabber board was mounted
on a Gateway 386/33 computer that was used to process the imaged data.
To reveal the signal, the background fluorescence was subtracted from
each frame. A cone-shaped spatial convolution kernel with a radius of 7
pixels (weighted average of neighboring pixels) was applied to improve
the signals through low-pass spatial filtering.7 32
Even
with this spatial averaging, the spatial resolution of this system is
under 1 mm.
|
Image Processing
Overview. Various image processing methods were used
to analyze the optical mapping "movies" of the cardiac
arrhythmias. These movies were composed of a series of 2-D images
acquired from the epicardial surface of the heart (which spans 3-D
space). It should be noted that we only present analysis of
experimental data derived from the 2-D manifestation of the underlying
3-D activity. Thus, we have described our experimental results in terms
of 2-D phenomena. For example, spiral waves and their cores are assumed
to be the surface manifestations of scroll waves and filaments,
respectively. Similarly, the rotation of these spirals is described as
being either clockwise or counterclockwise, which is ambiguous in three
dimensions (the 3-D analogy is torque).
Isochrone Maps.
Isochrone maps were generated from the
filtered video imaging data by analyzing the value of each pixel over
time. A point in each time series was labeled as part of a wave front
if it was the fastest part of the upstroke; that is, the maximum first
derivative. Activation thresholds helped to eliminate most maxima due
to noise. Because of motion-induced smearing, a set of pixels
perpendicular to the motion of the wave front activated in a single
frame (16.67-millisecond interval). Thus, the resulting wave fronts
seen in the image data were not lines but bands (see Figs 3 through
10).
Successive isochrone regions are separated by 16.7 milliseconds,
and the color scale indicates isochrone areas progressing in the
sequence red, yellow, green, blue, and purple.
|
|
|
|
|
|
|
|
Pseudo-ECGs. The spatial information obtained from the
optical mapping experiments may be displayed conveniently as temporal
patterns mimicking the ECG.1 7 19
Pseudo-ECG leads were
calculated as follows: (1) Each video frame was divided into two
halves, either vertically or horizontally. (2) At each point in time
(one video frame), the average transmembrane voltage activity (the
change in fluorescence intensity) obtained from all pixels in one half
of the frame was calculated. (3) The average value was also calculated
for the opposite half. (4) These two values then were subtracted from
each other according to the expression
Ex=
Eleft-Eright,
where Eleft and Eright represent,
respectively, the sum of the pixel values from the left and right
halves of the frame. The vertical (Ey) lead was calculated
in a similar manner. Although the pseudo-ECG is different from the
traditional ECG, it captures the important aspects of a true ECG. The
various ECGs recorded from the body surface exhibit various
morphologies due to their placement in relation to the heart. The
pseudo-ECG gives a similar measure of electrical activity by
calculating the difference of the transmembrane signal from 40 000
sites on the ventricular epicardium.
Time-Space Plots. Time-space plots (TSPs) show, in a single picture, the evolution of electrical activity over time for a given region of the heart.7 17 Two-dimensional image data from each frame were projected onto one line; lines from successive frames were stacked sequentially to form an image of time versus space. To create a horizontal TSP showing the evolution of activity in the horizontal direction, each column of pixels in the selected region was summed into a single pixel, resulting in a line. The lines from successive frames were stacked to create an image where the x axis corresponds to horizontal space (the same as the x axis of the selected region) and the y axis represents time (the duration of the series of frames); that is, to obtain a TSP T (with dimensions XxZ pixels) from a movie P consisting of Z frames, each of XxY pixels:
 | (1) |
For
a vertical TSP, the roles of the lines and columns were
switched; lines were summed to obtain single columns; the resulting
columns from each frame then were aligned successively. Spiral wave
parameters such as core size and position can be determined using TSPs.
Fig 2 shows the horizontal TSPs for computer simulations
of a single spiral wave (panel A) and a spiral wave pair "figure-of
eight" pattern (panel B). Notice that the horizontal core positions
are easily identifiable; the vertical core positions could similarly be
identified from the vertical TSP (for further details, see References 1
and 17).
|
Initiation of Arrhythmias. The rabbit heart
normally does
not undergo sustained VT at temperatures between 37° and 40°C. In
some experiments we were able to obtain spontaneous onset of sustained
VT when the temperature of the coronary perfusion solution was lowered
to 32°C. In most cases, however, lowering the temperature alone
was not sufficient to initiate VT. In such cases, we used one or two
pairs of bipolar electrodes located on the epicardial surface of either
ventricle to apply high frequency or programmed stimulation with a
strength of 1 to 2 V. With a single bipolar electrode, we first
attempted to induce tachycardias with the S1-S2
protocol, which consists of applying a single premature stimulus at
varying intervals following a basic stimulus train (basic cycle length,
300 to 400 milliseconds). If this failed to induce an arrhythmia, two
premature stimuli were applied
(S1-S2-S3 protocol). If the
single-electrode techniques failed, two electrodes were used following
the protocol used by Chen et al.9 This method of VT
initiation is based on the so-called "pinwheel" experiment
devised by Winfree.6 33 Briefly, a point stimulus
(S1) causes a wave to propagate with a roughly elliptical
(for an anisotropic medium) wave front. A second point stimulus
(S2) is applied at a certain distance from the first
stimulus point. This second stimulus causes cells in an elliptical
region to be depolarized. If the latter region overlaps with cells that
are refractory from S1, a wave break may occur. The
cells depolarized from the S2 stimulus will initiate wave
propagation with the important exception of those cells that are
absolutely refractory. These absolutely refractory cells will cause a
wave break resulting initially in figure-of-eight reentry.
Numerical Approach
Mathematical Model of Scroll Wave
Activity
Panfilov and Keener Model. We carried out
computer
simulations to facilitate insight into the interpretation of the
relation between the ECG and the underlying 3-D electrical activity in
the isolated heart. The Panfilov and Keener model incorporates a
realistic 3-D representation of left and right ventricular
geometry, with an accurate representation of myocardial fiber
distributions30 and a mathematical representation
of action potential dynamics.34 Recently, Nielsen et
al35 measured the geometry and fiber orientation in the
left and right ventricles of the intact canine heart to reconstruct the
data as a finite element model. Panfilov and Keener36 used
the data of Nielsen et al35 to construct an
electrophysiological model of the heart and to study the geometry of
wave propagation after stimulation of the apex as well as during
reentrant excitation. We used the same model but with isotropic
conditions to study the dynamics of scroll waves and how these scroll
waves are manifest in the simulated ECG. The excitable dynamics
equations used here are the linearized Fitzhugh-Nagumo
equations34 :
 | (2) |
 |
where
E is the excitation state variable (analogous to
transmembrane potential); g is the recovery state variable,
x, y, and z are the spatial
variables, and f(E)=-C1*E when
E<E1;
f(E)=C2*E+a when
E1
EE2;
f(E)=-C3*(E-1) when E>E2.
The parameter
values were C1=20, C2=3,
C3=15,
k=3. For the stationary scroll wave simulations,
E1=0.0065,
E2=0.841, a=0.15, and
(E)=1 when
E<E1; (E)=2 when
E1EE2; (E)=3
when
E>E2. For the nonstationary scroll wave simulations,
E1=0.0026, E2=0.837, a=0.06, and
(E)=1 when E<E1 and g<1.8;
(E)=2 when EE2;
(E)=3
when E>E2. With these parameter values, the function f(E)
was continuous. The dynamics of the recovery variable g in Equation
2
are determined by the function (E). In (E), the parameter
1 specifies the duration of the refractory tail and
3 specifies the duration of the excited state. Either of
these states can be lengthened by decreasing the corresponding value of
. For the stationary scroll wave simulations, we used the following
values: 1-1=1.0,
2-1=17.0,
3-1=1.0, D=4.0, and
dt=0.025; for the
nonstationary scroll wave simulations, we used the following values:
1-1=75.0,
2-1=1.0,
3-1=5.6, D=2.0, and
dt=0.05 (taken from the
meandering regime of the model of Panfilov and
Hodgeweg37 ). To establish a time scale, we scaled time
such that the period of reentry in the model was 150 milliseconds,
which corresponds to observed values of the period of reentry in the
rabbit heart. In these computations, the geometry data are mapped onto
a 3-D grid of 94x94x94 elements with 1-mm physical distance
between
grid points. Equation 2 was integrated using the Euler method
with
Neumann boundary conditions. The computer simulations were performed on
a Sun SPARCstation 10, model 512.
Obviously, this model is an oversimplified representation, a caricature, of the scroll wave dynamics that may be present during reentrant arrhythmias in a real heart. The only purpose of using this model is to guide our understanding of how wave propagation dynamics in three dimensions are manifest on the heart surface and in the ECG.
Calculation of the ECG. The simulated ECG was determined by calculating the dipole source density, based on the excitation variable E, for each element, assuming an infinite volume conductor.38 Three pairs of extracellular recording electrodes were placed orthogonally in the three axis directions, each 140 elements away from the center of the cellular matrix. The three orthogonal ECG leads were constructed by taking the difference in potential for each of the pairs.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Langendorff-Perfused Rabbit Heart
The video imaging approach is ideally suited to study the dynamics of vortexlike reentry in the isolated heart. Since 3-D electrical recordings are not available at sufficient resolution, we must analyze the surface manifestations of the underlying 3-D activity. By high-resolution mapping of the entire surface of the ventricle facing the video camera, we obtained images of the electrical activity that provided clues about the underlying mechanism.
Normal Sinus
Rhythm
The ECG during sinus rhythm (Fig 3A) shows a
basic
cycle length of 352 milliseconds, which is somewhat slow for the rabbit
(normal, 315 milliseconds); the P waves were biphasic but of normal
duration (30 milliseconds); the PR interval was 60 milliseconds (within
the normal range for the rabbit). Fig 3B shows a color
isochrone map of
the epicardial surface of the right atrium and the free wall of the
right ventricle during normal sinus rhythm. Temperature was 38°C. The
red and yellow (0 to 34 milliseconds) corresponds to atrial activation.
The first ventricular breakthrough occurred on the apex at about 67
milliseconds (blue). The wave front proceeded rapidly in the upward
direction to activate the majority of the right ventricle at 84 to 100
milliseconds (purple) and finally the ventricular base at 100 to 117
milliseconds (white). The total time of activation estimated by the
isochrone map corresponds well with the QRS duration from the ECG (60
milliseconds).
Spiral Wave Patterns and Tachycardia
Single Spiral Wave. Data obtained during a fast PVT
(period of 103±4 milliseconds) induced by
S1-S2 stimulation using two different point
sources (see "Methods") are shown in Fig 4. A
two-second horizontal ECG trace is presented in panel A. This trace
was not obtained simultaneously but within 30 seconds of the
accompanying optical recordings. An isochrone map of electrical
activity from the anterior ventricular surface during 100 milliseconds
is shown in panel B; in this panel, a spiral wave is clearly seen
rotating in a counterclockwise direction. The core of the spiral,
however, was at the edge of the field of view, so it was not possible
to determine if this vortex was part of a double vortex pair
(figure-of-eight pattern).
Figure-of-Eight Reentry. The data
presented in Fig 5 were recorded from another experiment in
which
programmed stimulation (S1-S2 pinwheel
protocol) at 34°C resulted in a long episode of PVT. Panel A shows
the horizontal and vertical pseudo-ECGs (see "Methods"). The mean
period of rotation was 152±11 milliseconds, and the QRS morphology was
irregular. A color isochrone map of the posterior epicardial
surface is shown in panel B. Two counter-rotating vortices
(figure-of-eight reentry) can be seen. Both cores are clearly visible
and are marked. As indicated by the arrows, the top vortex is rotating
in a counterclockwise manner and the bottom vortex is rotating in the
clockwise direction. A horizontal TSP from the first half of the same
episode is shown in panel C. Notice that initially there are two
identifiable cores in the TSP (panel C, top). Subsequently, only one
core can be distinguished, and then again, two cores can be seen. From
these data it may be concluded that this episode of PVT resulted from
nonstationary reentry manifest on the anterior free wall as a pair of
counter-rotating vortices (panel B) whose organizing centers (cores)
were not stationary.
The data in Fig 6 were obtained
from another preparation
after sustained VT was induced by the application of an
S1-S2-S3 protocol (basic cycle
length, 400 milliseconds) at 32°C. Panel A shows the horizontal ECG
signal that was low-pass filtered at 40 Hz to remove noise artifacts.
The RR interval was 150±6 milliseconds, indicating a stable period.
Although the fluorescence signal was less than optimal in this
experiment, we were able to map and locate the origin of the activity
for several beats. Panel B is a color isochrone map of the free wall of
the left ventricular surface. A rotor was established near the apex
(rotor 1), which gave rise to rapid excitation of the entire heart.
Evidence for the coexistence of another rotor (rotor 2) somewhere in
the posterior wall is apparent near the left side of the figure (orange
and yellow isochrones). The wave front emerging from the left side
(from the right ventricle) cannot be due to rotor 1 because if the
activity were to wrap around the heart from rotor 1, the activation
sequence there would be counterclockwise, not clockwise as observed.
The ECG pattern corresponds to a sustained VT with the typical wide and
aberrant morphology expected from that generated by two
counter-rotating but synchronized vortices, one manifested on the
anterior and the other on the posterior ventricular wall surfaces. As
shown by the ECG, the QRS amplitude changed on a beat-to-beat basis,
which suggested that there might be recurrent conduction abnormalities
throughout the episode. In fact, as shown in Fig 6C by an
individual
pixel recording of local activity during reentry, there was a small
region on the ventricular epicardial surface that showed 2:1
activation. Alternatively, the polymorphic shape of the reentrant
tachycardia might have been caused by beat-to-beat changes in the
position of either one or both spiral cores. We were able to identify
several core positions from rotor 1 using the TSPs. Movement of the
core was apparent, but we could not link the core position with the
beat-to-beat changes in QRS morphology. There was very little movement
of the core in the horizontal direction, but the core moved in the
vertical direction over the duration of the experiment and it remained
near the apex. The vertical movement of the core may not have been
apparent in the horizontal ECG. The lack of correspondence between the
core positions and QRS morphology is probably due to a combination of
the region of 2:1 activation and the three dimensionality of the scroll
wave dynamics.
Nonstationary Spiral and Electrical Alternans. In Fig
7, we show data from an experiment at 30°C. These data
were obtained about 30 minutes after successful induction of an
arrhythmia by programmed S1-S2 stimulation. The
period was longer (200±15 milliseconds) than in the previous examples,
and the pseudo-ECGs showed alternans in the QRS amplitude as well as
slower undulations with a period of 12 beats (panel A). The slow
undulations were due to meandering of the organizing center of the
reentrant activity as determined by localization of the core with TSPs.
The core position remained within a small region but was not
stationary, and the distance between extreme core positions was
approximately 3 mm. On the other hand, detailed analysis of the
optical recordings demonstrated that the beat-to-beat alternans in ECG
amplitude was due to 2:1 block in the basal portion of the heart (see
below). As demonstrated by the color isochrone maps in panels B and C,
this experiment demonstrates two types of surface manifestations
expected from scroll wave activity in three dimensions, spiral waves,
and breakthrough patterns.21 Panel B shows an isochrone
map that displays a breakthrough pattern in beat 14. Panel C shows the
isochrone map just two beats later (beat 16), when a spiral wave
reentrant pattern was evident.
Individual pixel recordings and a
vertical TSP illustrating the 2:1
block in the basal region of the heart are shown in Fig 8. From
the vertical TSP (Fig 8B), a spiral wave core
can be identified in beat 16, and the corresponding isochrone map (Fig
7C) shows a spiral wave pattern. No spiral wave is evident for
beat 14,
however, when the breakthrough pattern was manifest on the surface (Fig
7B). The pixel recording from the base of the heart (Fig
8C) showed 2:1
block. The pixel recording from the midventricle (Fig 8D)
showed
activation at the same rate of the arrhythmia (200 milliseconds). The
individual pixel recording for the point identified as the core for
beat 16 is shown in Fig 8E. Notice the undulating pattern in
the
individual pixel recording, indicating that the core meanders in this
area. The signal recorded from a nearby point (5 mm away), however, did
not show this undulating pattern, indicating that the core remained in
a relatively small area. Such a small degree of meandering is reflected
in the pseudo-ECG as small but appreciable slow undulating changes in
the QRS morphology superimposed on the 2:1 alternans.
Meandering Spiral Wave. The application of a
high-frequency
(50 Hz) burst of 50 pulses (duration, 10 milliseconds; strength, 10 V)
produced the arrhythmia shown in Fig 9. The horizontal
pseudo-ECG (Ex) and the true ECG are shown together
in panel A for comparison and demonstrate excellent agreement. The ECGs
show continuously changing morphology and an irregular period. The rate
of this arrhythmia was difficult to determine from the ECGs; however,
spectral analysis of the true ECG (not shown) exhibited a peak near
8 Hz, which corresponded to a dominant period of 125 milliseconds. A
spiral wave was initially apparent on the surface of the
ventricle (panel B) but then drifted out of the field of view only to
return at the end of the 2-second episode. During this episode, the
organizing center of the spiral wave moved over much of the ventricular
surface in the field of view. The fact that it reappeared indicates
that this spiral wave may have moved over the entire ventricular
epicardium. The distance between the two extreme core positions within
the field of view was 30 mm, as determined from both horizontal and
vertical TSPs.
Torsade de Pointes. The data presented
in Fig 10 show another example of a sustained but
nonstationary reentrant arrhythmia (period, 155±12 milliseconds). The
pseudo-ECG recordings (Ex and Ey) demonstrated
that this was an episode of PVT. The typical pattern of torsade de
pointes, characterized by undulations in the QRS morphology, can be
clearly seen. In the bottom panel, frames a through g are sequential
color isochrone maps from the anterior wall of the ventricular surface,
each of which corresponds to a single QRS complex in the pseudo-ECG, as
indicated by the respective letters and arrows in the top panel. These
maps demonstrate the direct correlation that exists between the spatial
patterns of vortexlike activity resulting from a nonstationary rotor
and the temporal patterns emerging in the pseudo-ECG as a result of the
movement. Map a (217 to 367 milliseconds) was obtained during beat a
with time zero (red) being arbitrarily chosen. Here we encounter a
complete counterclockwise rotation (curved arrow) of a highly organized
vortex around a small core. Although most of the epicardial wall facing
the video camera was activated directly by the rotating wave, apex
activation occurred rapidly from the posterior wall (upwardly directed
arrow). In map b (367 to 517 milliseconds), the rotor had moved toward
the left and the apex activation became somewhat delayed. This shift
coincided with a change in the morphology of the QRS. Subsequently, in
map c (517 to 667 milliseconds), the core became even more elongated
and shifted downward, which again resulted in a change in the QRS
morphology. In map d (684 to 834 milliseconds), the spiral pattern was
interrupted by functional block near the apex due to the late
depolarization of this area in the previous beat (see map c). In the
next beat, the wave front appeared to wrap around the right ventricle
onto the posterior wall, to return through the apex and initiate a new
rotation (map e, 834 to 984 milliseconds), starting with the elongated
diagonal area (red) on the center of the anterior wall. This was
manifest as a large decrease in the amplitude of the QRS in both
pseudo-ECGs. Once again, a functional block occurred in a region of
late depolarization in the previous beat. In map f (1000 to 1150
milliseconds), the area of functional block from the previous beat
became excitable and the activity propagated into this region.
Meanwhile, the activity had continued along its main reentrant pathway.
These two waves of activity collided, and this collision acted to
"reset" the pattern of activation, as shown in map g (1150 to
1300 milliseconds), which is very similar to map a. The activation from
the apex occurred in a 1:1 manner, with the reentrant wave suggesting
that both waves originated from a single source. Clearly, although
highly complex, the spatial patterns manifest on the epicardial surface
that accompany the undulating ECG signal resembling torsade de pointes
appear to result from the gradual migration and deformation of the
organizing center (the core) of the reentrant activity as well as
collision of wave fronts. The movement of the core (when it appeared)
was systematically localized using both horizontal and vertical TSPs of
small areas of the image area. The positions of the core are shown in
Fig 10 as asterisks. The gradually changing QRS morphology
after QRS
complex g was similar to that following complex a, with the exception
that the transition was slower. The movement of the core in beats g
through i was similar to the movement in beats a through c. In beats g
through i, however, the core traversed less space (in the same amount
of time) as compared with beats a through c. Therefore, the velocity of
core movement corresponds well to the undulations in the
pseudo-ECGx (top panel in Fig 10). The points
identified as
spiral wave cores showed undulating patterns in the individual pixel
recordings. As a matter of fact, many of the pixels showed an
undulating pattern, indicating that the spiral core was moving over
much of the heart surface.
Scroll Wave Movement and ECG
Polymorphism
In an effort to provide an analysis of association of
scroll
wave movement to polymorphism in the ECG signal, we calculated the
extreme core positions for five of the six episodes described above
(only one core position was identified from the data presented in
Fig 4, therefore extreme positions could not be identified).
Core
positions were identified using the horizontal and vertical TSPs as
described above. The distance between the extreme core positions
identified in the recordings presented in Figs 7 and
8 was
approximately 3 mm. The VT episode shown in Fig 6 exhibited
somewhat
larger variations in QRS amplitude in the ECG, and the distance between
the extreme core positions was 7 mm. The episode shown in Fig 5
was a
figure-of-eight reentry, therefore two core positions could be
followed. The distances between the extreme core positions for these
two cores were 4 mm and 6 mm. The fact that these two core positions
did not move together (see Fig 5C) may have led to the high
degree of polymorphism in the ECG pattern. Furthermore, for the
episodes shown in Figs 9 and 10, large
variations in QRS amplitude were
observed, and the distances between the extreme core positions were 30
and 15 mm, respectively. The VT episode presented in Figs 7 and
8
showed alternans in QRS amplitude coexisting with slow undulations
throughout the episode. The alternans was shown to have occurred
because of functional block near the base, whereas the small-amplitude
undulations were the result of small movements (meandering) of the
core. In the experiment immediately after this episode, the core became
stationary, and the pseudo-ECG for this stationary episode was very
similar to that for the nonstationary episode (Figs 7 and
8) with the
exception that no slow undulations were observed, indicating that the
undulations were in fact due to small movement of the spiral core.
Therefore, the distance between the extreme core positions ranged from
3 mm to 30 mm, and the episodes with the largest distances between
extreme positions showed the greatest variation in ECG morphology.
Model Results
Currently available video imaging technology
does not allow the
means to determine whether 3-D scroll waves are indeed the mechanism
underlying arrhythmias in the structurally normal heart. The data
presented above for the Langendorff-perfused rabbit heart were from
the surface of the heart only (a layer of epicardium 400 µm thick
contributes to the fluorescence signal). We therefore carried out
numerical experiments using the geometrically realistic Panfilov and
Keener model (see "Methods") to demonstrate how known 3-D scroll
wave patterns of activity are manifest on the 2-D surfaces and in the
ECG. A scroll wave was initiated by setting a small region of cells to
the excited state throughout the myocardium of the left ventricle from
the base to halfway down the heart and setting a region of cells
immediately adjacent to these excited cells to the absolute refractory
state. This caused the wave to wrap around the refractory cells and
generate sustained reentrant activity. We simulated both stationary and
nonstationary scroll waves as described above.
The heart geometry and
the coordinate system axes are shown in panel A
of Fig 11. The scroll wave that was initiated with the
initial set of parameters remained stationary. The scroll wave remained
at the site of initiation in the middle of the left ventricle. A
snapshot of epicardial activity on the left ventricle of this
stationary scroll wave is shown in panel B of Fig 11, with
white
representing excited tissue and red representing heart
tissue that was not excited. This spiral wave evident on the surface of
the myocardium is only the surface manifestation of the underlying
reentrant 3-D scroll wave shown in panel C. The linear scroll wave
filament spanned the entire left ventricular wall and gave rise to a
monomorphic pattern of tachycardia in the three
orthogonal-lead–computed ECGs. The y-lead ECG is shown in panel
D.
|
The results from the simulations involving the nonstationary scroll
wave set of parameters are shown in Fig 12. In these
computer simulations, the parameters were such that irregular
meandering occurred at a velocity 40% of the wave propagation
speed. Panels A and B show isochrone maps from two beats starting at
times t1 and t2, respectively. Notice
that the epicardial surface manifestation (reentrant spiral wave) was
nonstationary and moved from near the apex at time=t1 to
the middle of the left ventricular wall at time=t2. The 3-D
scroll waves at times t1 and t2 are shown in
panels C and D, respectively. Notice that the scroll wave filament
spans the entire myocardium, and the filament movement is the cause of
the nonstationary epicardial spiral wave patterns. The meandering of
the scroll wave gave rise to undulating patterns characteristic of
torsade de pointes in all three ECG leads (panel E).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The results presented here demonstrate that PVTs induced in the structurally normal, isolated rabbit heart are the result of nonstationary vortexlike reentrant activity. Indeed, high-resolution video imaging revealed that reentrant patterns are manifest on the ventricular epicardial surface as nonstationary spiral waves, which activate the ventricles at high frequencies and may lead to various patterns of localized functional block, even in areas that are remote from the core of the spiral. Since in these experiments the entire ventricular wall remained viable as a result of coronary perfusion, it is reasonable to expect that such spirals are in fact the 2-D representation of 3-D scroll waves spanning the thickness of the wall.
Relevance of the Numerical and Experimental Approaches
We
used two simplifying approaches to study the mechanism of
polymorphic ventricular arrhythmias. First, we used high-resolution
optical mapping of the epicardial transmembrane potential in the
intact, healthy, isolated, Langendorff-perfused rabbit heart. We
believe that studying the normal heart (in the absence of ischemia,
necrosis, neural reflexes, and so forth) is a necessary step in
understanding the more complicated arrhythmias associated with cardiac
disease. We believe that because our preparations were isolated,
perfused, and temperature was low and controlled, we were able to
obtain arrhythmias that were stable, not transient, which is consistent
with results obtained by Bardy et al.39 In contrast, in
the experimental open chest protocols (with the exception of
cardiopulmonary bypass), arrhythmias produce drastic changes in the
cardiopulmonary hemodynamics, with rapid alteration of the condition of
the heart (for example, blood pressure and ejection fraction). Our
second approach to investigate PVT was to use a mathematical model of
the whole heart based on the theory of generic excitable media. The
theory of wave propagation in excitable media is applicable to the
study of the dynamics and mechanisms of electrical impulse propagation
in cardiac tissue because the electrical activity in the heart exhibits
many of the characteristics of generic excitable
media.19 33
Spatial and Temporal Resolution
Traditional extracellular
recordings are limited by the low
spatial resolution and by recording from the extracellular space.
Although our high spatial resolution system has low temporal resolution
(sampling rate is 16.67 milliseconds), the areas where high spatial
resolution is required (near the phase singularity or core) have slow
conduction,6 33 so the slow sampling rate is not a
problem. In addition, although the sampling rate is 16.67 milliseconds,
the camera shutter is open for this entire interval. Therefore, the
fluorescence is integrated in this 16.67-millisecond interval, which
acts to smooth high-frequency information and thus prevent aliasing.
The period of the tachycardias in our preparations ranges from 100 to
200 milliseconds; this allows us to create at least five isochrones for
each beat. The epifluorescence measured in our system reflects changes
in transmembrane potential that are easier to interpret and allow
higher spatial localization compared with extracellular electrode
recordings. This high spatial resolution spanning the entire
ventricular surface allowed us to identify reentrant circuits with less
than 1-mm resolution.
2-D Manifestations of 3-D Scroll Waves
From our experiments,
it is not possible to discern the 3-D
patterns of electrical activity throughout the myocardium. To address
this issue, we performed numerical simulations to investigate how
scroll waves are manifest in the ECG and in the activity patterns on
the epicardial surface. The possibility of transmural propagation,
however, must be considered. It has been shown that on the epicardial
surface of the Langendorff-perfused rabbit heart, the smallest
wavelength (conduction velocity times refractory period) occurs in the
transverse direction at a pacing interval of 75 milliseconds and is
approximately 1 cm.40 The rabbit ventricular wall is
considerably thinner than 1 cm; therefore, the electrical activity
throughout the myocardium would be expected to be similar to the
epicardial surface manifestation, with only slight modifications due to
rotational anisotropy and the Purkinje fibers in the subendocardium. In
addition, all of the data presented in this study demonstrate
continuous wave propagation on the surface of the myocardium resulting
in continuous isochrone bands (see Figs 4 through
10),
suggesting that
transmural activation did not occur during our recordings. Also, the
surface patterns were spiral waves, which can only result from
reentrant processes. These factors strongly argue that the surface
recordings presented above accurately reflect the activity
throughout the myocardial wall.
Epicardial Stimulation and Formation of Scroll Waves
It is
important to note that our stimulation protocols (epicardial
surface stimulation) may have favored the formation of nonintramural
reentrant circuits on the ventricular epicardial
surface.41 Similar epicardial stimulation protocols used
by Chen et al9 and Frazier et al10 resulted
in scroll wave filaments that appeared to stretch from the epicardium
to the endocardium, thus giving rise to transmural reentrant circuits
comparable to the data presented above. The placement and strength
of the stimulation affects the formation of the initial filament
shape.21 For large stimulus strengths, the scroll wave
filament spans the entire myocardium, giving rise to transmural
reentrant circuits that manifest as spiral waves on both the epicardial
and endocardial surfaces. If the stimulus strength is weaker, however,
the filament does not span the entire myocardium, and although
reentrant spiral waves are apparent on the epicardium, only
breakthrough patterns would be expected to occur on the endocardium. It
was not possible to determine the shape of the vortex filament in our
experiments; therefore, it is premature to speculate on whether the
filaments spanned the entire myocardium. However, due to the reentrant
spiral wave patterns observed on the epicardium, we can clearly rule
out the possibility of intramural reentry. As discussed by
Winfree,41 other stimulation protocols involving
endocardial stimulation are expected to give rise to intramural
reentrant circuits similar to those observed by Pogwizd and
Corr42 in which only breakthrough patterns are evident on
the heart surfaces.
Nonstationary Scroll Waves in the Heart
We have proposed that
reentrant excitation in the isolated rabbit
heart is analogous to scroll waves in other excitable media. The
following observations argue in favor of this hypothesis: (1) Sustained
reentry may be consistently generated by point stimulation at high
frequency or by programmed stimulation using the pinwheel
protocol.6 33 In fact, the characteristics of the
stimulus
parameters used to initiate the activity were derived from principles
that govern the induction of "wave breaks" with consequent spiral
wave formation in 2-D and 3-D media.7 9 (2) Reentrant
activity appears to revolve around an elongated organizing center (a
rotor) whose extension appears to be smaller17 than
previously thought. In 2-D media, such a rotor is thought to be a point
or a line, depending on the excitability of the medium, whereas in
three dimensions, it is known as a filament.33 Such an
organizing center is functionally determined and is thought to be the
result of "curling" of the wave front. In fact, depending on the
excitability of the medium, its size is determined by the curvature of
the wave front18 and the refractory period of the
medium.6 33 (3) The organizing cores of the vortices
were
not stationary, which is a property of rotors in a wide variety of
excitable media, particularly in the presence of heterogeneities or
parameter gradients. It is reasonable to expect that gradients in the
structural and electrophysiological43 properties of the
ventricular muscle of the rabbit heart may have contributed to the
dynamics of the spiral movement in our experiments. However, in the
absence of direct evidence for such gradients, it is premature to
provide an explanation. Nevertheless, the theory of generic excitable
media allows us to speculate further on the mechanisms causing the
movement. It has been shown that drifting of spiral and scroll waves in
generic excitable media occurs due to parameter gradients such as
action potential duration and conduction velocity44 as
well as boundary curvature45 and wall
thickness.46 In addition, the shape of the scroll wave
filament alone may be the cause of drift.34 Finally, the
scroll wave movement is influenced by the specific properties of the
excitable media. Even in the absence of heterogeneities, the movement
of spiral waves can be complex, as shown in Fig 12. Varying
the
parameters in even the simple Fitzhugh-Nagumo equations results in
movement of spiral waves in circular paths, flower patterns
(meandering), and more complex paths (hypermeandering).47
Most likely, the complex movement of the core that we observed is due
to a combination of these factors. The unique features of the scroll
wave concept are that it accounts for the 3-D nature of the heart, the
effects of wave front curvature,48 and the movement of the
organizing center and therefore differs from the leading circle concept
of functional24 and anisotropic49
reentry.
Breakthrough Patterns
Breakthrough patterns of activation on
the heart surface are
generally thought to be associated with focal sources. These
breakthrough patterns, however, may be the result of 3-D reentry within
the myocardium. Computer simulations have shown that breakthrough
patterns and/or spiral wave patterns including figure-of-eight patterns
result on the surface from underlying 3-D scroll waves, depending on
the initiation protocol and consequently on the shape of the
filament.21 In fact, in one of our experiments a
breakthrough pattern was observed on the surface of the ventricle (Fig
7B), and a few beats later a spiral wave was apparent (Fig
7C) and no
major changes were observed in the ECG pattern (Fig 7A). The
moving
breakthrough patterns observed by Bardy39 that coincided
with an undulating ECG similar to torsade de pointes may have been the
result of a drifting scroll wave within the myocardium.
Torsade de Pointes and Long QT
Torsade de pointes was
originally described based solely on the
periodic undulations in QRS morphology observed in the
ECG,4 although it is commonly associated with a long QT
interval.5 Today, there is wide agreement among
investigators that torsade de pointes that develops in individuals
presenting long QT intervals is the result of triggered activity
secondary to early afterdepolarizations (EADs). However, although EADs
are potentially a viable mechanism for the initiation of this type of
arrhythmia and there is experimental evidence linking EADs to the onset
of polymorphic tachycardias,2 it is difficult to relate
this mechanism to the characteristic undulating ECG pattern observed
during the episodes of torsade. As demonstrated by our experiments
and computer simulations, the dynamics of meandering or drifting spiral
waves recorded on the ventricular epicardial surface correlate directly
with ECG patterns that are very similar to those of torsade de pointes.
Our experimental results (Figs 9 and 10) showed
that an undulating ECG
pattern was associated with a nonstationary reentrant vortex that moved
over a large portion of the ventricular surface. The computer
simulations (Fig 12) demonstrated that meandering scroll waves
in an
isotropic, homogeneous model in a realistic heart geometry give rise to
periodic undulations in the ECG characteristic of torsade de pointes.
Finally, a long QT interval denotes a long action potential duration
(APD), which is not only related to the development of EADs but may
have potential implications concerning scroll wave dynamics as well, at
least in principle. In fact, it has been demonstrated in computer
simulations that as the APD in the Fitzhugh-Nagumo model is increased
(by decreasing in Equation 2), the motion of the rotor
core becomes
more complex.47 Hence, although there are many factors
involved, merely increasing APD may create the dynamics necessary to
induce the complicated scroll wave dynamics that we believe are
responsible for PVT. Our results confirm Winfree's prediction that
torsade de pointes results from meandering scroll waves in the
myocardium.6
Scroll Waves and PVT
There is considerable evidence from our
results that PVTs in the
isolated rabbit heart are due to nonstationary 3-D scroll waves. We
observed changes in the surface manifestations during PVT and were able
to calculate the position of the core of spiral waves observed on the
heart surface. During PVT, the core positions were not stationary. In
fact, the degree of nonstationarity appeared to correspond (at least
qualitatively) with the degree of polymorphism. The direction of the
spiral wave movement most probably plays a role in the ECG patterns,
which requires further study. The core movement for the VT episode in
Fig 6 was mostly in the vertical direction, while the core
movement in Figs 9 and 10 was both horizontal
and vertical. This may
explain the greater degree of undulations in the ECG of the latter
examples. Another piece of evidence that PVTs result from nonstationary
scroll waves is that individual pixel recordings from the surface of
the heart showed a reduced amplitude when the core was moving nearby,
resulting in undulating patterns (Fig 8E). It has been shown
using
microelectrodes that the membrane potential oscillations near the core
of reentrant activity show a reduced amplitude.7 50
Slow
conduction corresponding to converging isochrones with steep wave
fronts is expected near spiral wave cores. In fact, we see slow
conduction and steep wave front curvature in the isochrone maps
presented in Figs 4 through
10.
Nonstationary scroll waves gave
rise to changes in the QRS morphology in the ECG and left complicated
patterns of refractory tissue in their wake. The fact that PVT often
degenerates into fibrillation51 is most probably due to
the hazardous consequence of nonuniform refractoriness resulting from
nonstationary scroll waves. These patterns of nonuniform refractory
tissue are causes of functional block, which in our experiments allowed
waves to propagate into these areas immediately after they became
excitable (see Fig 10f). In conclusion, the only mechanism
that
explains the patterns of electrical activity that we recorded on the
epicardial surface (spiral waves, figure-of-eight reentrant patterns,
and breakthrough points) during PVT in the isolated rabbit heart is
nonstationary scroll waves. Although there are many differences between
cardiac arrhythmias in the isolated rabbit heart and those occurring in
humans, our data suggest that moving scroll waves are one possible
mechanism for PVT in patients with structurally normal hearts.
|
Acknowledgments |
|---|
This study was supported in part by grants PO1-HL39707, RO1-HL29439, and RO1-HL46148 from the National Heart, Lung, and Blood Institute, National Institutes of Health. Dr Davidenko is an Established Investigator of the American Heart Association. We would like to thank JoAnne Getchonis, Christine Kapuscinski, and Wanda Coombs for technical assistance.
Received October 13, 1994; accepted November 26, 1994.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Davidenko JM. Spiral wave activity: a possible common mechanism for polymorphic and monomorphic ventricular tachycardias. J Cardiovasc Electrophysiol. 1993;4:730-746. [Medline] [Order article via Infotrieve]
2. Jackman WM, Szabo B, Friday KJ, Margolis PD, Moulton K, Wang X, Patterson E, Lazzara R. Ventricular arrhythmias related to early after depolarizations and triggered firing: relationship to QT interval prolongation and potential therapeutic role for calcium channel blocking agents. J Cardiovasc Electrophysiol. 1990;1:170-195.
3.
Smirk FH, Ng J. Cardiac ballet repetitions of complex
electrocardiographic patterns. Br Heart J. 1969;31:426-431.
4. Dessertenne F. La tachycardie ventriculaire à deux foyers opposés variable. Arch Mal Coer. 1966;56:263-272.
5. Jackman WM, Friday KJ, Anderson JL, Aliot EM, Clark M, Lazzara R. The long QTR syndromes: a critical review, new clinical observations and a unifying hypothesis. Prog Cardiovasc Dis. 1988;31:115-172. [Medline] [Order article via Infotrieve]
6. Winfree AT. Electrical instability in cardiac muscle: phase singularities and rotors. J Theor Biol. 1989;138:353-405. [Medline] [Order article via Infotrieve]
7.
Pertsov AM, Davidenko JM, Salomonsz R, Baxter WT, Jalife J.
Spiral waves of excitation underlie reentrant activity in isolated
cardiac muscle. Circ Res. 1993;72:631-650.
8.
El-Sherif N, Mehra R, Gough WB, Zeiler RH. Ventricular
activation patterns of spontaneous and induced ventricular rhythms in
canine one-day-old myocardial infarction. Circ
Res. 1982;51:152-166.
9.
Chen PS, Wolf PD, Dixon EG, Danieley ND, Frazier DW, Smith
WM, Ideker RE. Mechanism of ventricular vulnerability to single
premature stimuli in open-chest dogs. Circ Res. 1988;62:1191-1209.
10. Frazier DW, Wolf PD, Wharton JM, Tang ASL, Smith WM, Ideker RE. Stimulus-induced critical point: mechanism for the electrical initiation of reentry in normal canine myocardium. J Clin Invest. 1989;83:1039-1052.
11. Inoue H, Murakawa Y, Toda I, Nozaki A, Mashima S, Sugimoto T. Epicardial activation pattern of torsades de pointes in canine hearts with quinidine-induced long QT U interval but without myocardial infarction. Am Heart J. 1986;111:1080-1087. [Medline] [Order article via Infotrieve]
12.
de Bakker JM, van Capelle FJ, Janse MJ, Wilde A, Coronel R,
Becker A, Dingemans K, van Hemel H, Hauer R. Reentry as a cause of
ventricular tachycardia in patients with chronic ischemic heart
disease: electrophysiologic and anatomic correlation.
Circulation. 1988;77:589-606.
13. Nhon N, Hope RR, Kabell G, Scherlag BJ, Lazzara R. Torsades de pointes: electrophysiology of atypical ventricular tachycardia. Am J Cardiol. 1980;45:494-503. Abstract.
14. Miller JM. The many manifestations of ventricular tachycardia. J Cardiovasc Electrophysiol. 1992;3:88-107.
15. Downar E, Harris L, Mickleborough LL, Shaikh N, Parson ID. Endocardial mapping of ventricular tachycardia in the intact human ventricle: evidence for reentrant mechanisms. J Am Coll Cardiol. 1988;11:783-791. [Abstract]
16. Kuck KH, Schulter M, Kunze KP, Geiger M. Pleomorphic ventricular tachycardia: demonstration of conduction reversal within the reentry circuit. PACE. 1989;12:1055-1064.
17. Davidenko JM, Pertsov AM, Salomonsz R, Baxter WT, Jalife J. Stationary and drifting spiral waves of excitation in isolated cardiac muscle. Nature. 1991;355:349-351.
18. Zykov VS. Simulation of Wave Processes in Excitable Media. New York/Manchester: University Press; 1987.
19. Jalife J, Davidenko JM. Spiral waves as a mechanism of reentrant excitation in isolated cardiac muscle. In: Shenasa M, Borggrefe M, Breithardt G, eds. Cardiac Mapping. Mt Kisko, NY: Futura Publishing; 1993:607-623.
20.
Winfree AT. Scroll-shaped waves of chemical activity in three
dimensions. Science. 1973;181:937-939.
21. Pertsov AM, Jalife J. Three-dimensional vortex-like reentry. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia: WB Saunders Co; 1995:403-410.
22. Keener JP, Tyson JJ. The dynamics of scroll waves in excitable media. SIAM Rev. 1992;34:1-39.
23. Winfree AT. Stable particle-like solutions to the nonlinear wave equations of three-dimensional excitable media. SIAM Rev. 1990;32:1-53.
24.
Allessie MA, Bonke FIM, Schopman FJC. Circus movement in
rabbit atrial muscle as a mechanism of tachycardia.
Circ Res. 1973;33:54-62.
25. Downar E, Parson ID, Mickleborough LL, Yao LC, Cameron DA, Waxman MB. On-line epicardial mapping of intraoperative ventricular arrhythmias: initial clinical experience. J Am Coll Cardiol. 1984;4:703-714. [Abstract]
26.
Janse MJ, Van Capelle FJL, Morsink H, Kleber AG, Wilm-Schopman
F, Cardinal R, D'Alnoncourt CN, Durrer D. Flow of `injury'
current
and patterns of excitation during early ventricular arrhythmias in
acute regional myocardial ischemia in isolated porcine and canine
hearts. Circ Res. 1980;47:151-165.
27. Cardinal R, Savard P, Carson L, Perry JB, Page P. Mapping of ventricular tachycardia induced by programmed stimulation in canine preparations of myocardial infarction. Circulation. 1984;1:136-148.
28.
Lin JL, Wilber DJ, Du D, Pearlman J, Ruskin JN, Garan H.
Localization of breakthrough site of canine monomorphic ventricular
tachycardia by pacemapping. Circulation. 1991;84:1319-1332.
29. Winfree AT. Theory of spirals. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia: WB Saunders Co; 1995:379-389.
30. Panfilov AV, Keener JP. Modeling reentry in a finite element model of the heart. J Physiol (Lond). 1993;467:152P. Abstract.
31.
Liu Y, Cabo C, Salomonsz R, Delmar M, Davidenko J, Jalife J.
Effects of diacetyl monoxime on the electrical properties of sheep and
guinea pig ventricular muscle. Cardiovasc Res. 1993;27:1991-1997.
32. Russ JC. The Image Processing Handbook. Boca Raton, Fla: CRC Press; 1992:53-74.
33. Winfree AT. When Time Breaks Down. Princeton, NJ: Princeton University Press; 1987.
34. Panfilov AV, Pertsov AM. Vortex rings in a three-dimensional medium described by reaction-diffusion equations. Dokl Biophysics. 1984;274:58-60.
35. Nielsen PMF, Le Grice IJ, Smaill BH, Hunter PJ. Mathematical model of geometry and fibrous structure of the heart. Am J Physiol. 1991:H1365-H1378.
36. Panfilov AV, Keener JP. Reentry in an anatomical model of the heart. Chaos, Solitons and Fractals. In press.
37. Panfilov AV, Hogeweg P. Spiral wave breakup in a modified Fitzhugh-Nagumo model. Phys Lett A. 1993;295-299.
38. Plonsey R, Barr RC. Bioelectricity. New York: Plenum Press; 1988:213-216.
39.
Bardy GH, Ungerleider RM, Smith WM, Ideker RI. A mechanism of
torsade de pointes in a canine model. Circulation. 1983;67:52-59.
40. Schalij MJ, Lammers W, Pensma PL, Allessie MA. Anisotropic conduction and reentry in perfused epicardium of rabbit left ventricle. Am J Physiol. 1992;32:H1466-H1478.
41. Winfree AT. Ventricular reentry in three dimensions. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 1st ed. Philadelphia: WB Saunders Co: 1990.
42.
Pogwizd SM, Corr PB. Reentrant and nonreentrant mechanisms
contribute to arrhythmogenesis during early myocardial ischemia:
results using three-dimensional mapping. Circ Res. 1987;61:352-371.
43.
Rosenbaum DS, Kaplan DT, Kanay A, Jackson L, Garan H, Cohen
RJ, Salama G. Repolarization inhomogeneities in ventricular myocardium
change dynamically with abrupt cycle length shortening.
Circulation. 1991;84:1333-1345.
44. Davidenko JM, Pertsov AM, Baxter WT, Salomonsz, Cabo C, Jalife J. Mechanisms of drift of spiral wave reentry in isolated epicardial muscle. Circulation. 1993;88(suppl 4):I-327. Abstract.
45. Davydov VA, Zykov VS. Kinematics of spiral waves on nonuniformly curved surfaces. Physica D. 1991;49:71-74.
46. Panfilov AV, Aliev RR, Mushinsky AV. An integral invarient for scroll rings in a reaction-diffusion system. Physica D. 1987;36:181-188.
47. Winfree AT. Varieties of spiral wave behavior: an experimentalist's approach to the theory of excitable media. Chaos. 1991;1:303-334. [Medline] [Order article via Infotrieve]
48.
Cabo C, Pertsov AM, Baxter WT, Davidenko JM, Gray RA, Jalife
J. Wave front curvature as a cause of slow conduction and block in
isolated cardiac muscle. Circ Res. 1994;75:1014-1028.
49. Wit AL, Dillon SM, Coromilas J, Saltman AE, Waldecker B. Anisotropic reentry in the epicardial border zone of myocardial infarcts. Ann N Y Acad Sci. 1990;591:86-108. [Medline] [Order article via Infotrieve]
50.
Allessie MA, Bonke FIM, Schopman FJC. Circus movement in
rabbit atrial muscle as a mechanism of tachycardia, II: the role of
nonuniform recovery of excitability in the occurrence of unidirectional
block as studied with multiple microelectrodes. Circ
Res. 1976;39:168-177.
51. Adhar GC, Larson LW, Bardy GH, Greene HL. Sustained ventricular arrhythmias: differences between survivors of cardiac arrest and patients with recurrent sustained ventricular tachycardia. J Am Coll Cardiol. 1988;12:159-165.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  L. Tang, G.-S. Hwang, H. Hayashi, J. Song, M. Ogawa, K. Kobayashi, B. Joung, H. S. Karagueuzian, P.-S. Chen, and S.-F. Lin Intracellular calcium dynamics at the core of endocardial stationary spiral waves in Langendorff-perfused rabbit hearts Am J Physiol Heart Circ Physiol, July 1, 2008; 295(1): H297 - H304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. B. Huffaker, J. N. Weiss, and B. Kogan Effects of early afterdepolarizations on reentry in cardiac tissue: a simulation study Am J Physiol Heart Circ Physiol, June 1, 2007; 292(6): H3089 - H3102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. V. Panfilov, R. H. Keldermann, and M. P. Nash Drift and breakup of spiral waves in reaction diffusion mechanics systems PNAS, May 8, 2007; 104(19): 7922 - 7926. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. F. Mironov, F. J. Vetter, and A. M. Pertsov Fluorescence imaging of cardiac propagation: spectral properties and filtering of optical action potentials Am J Physiol Heart Circ Physiol, July 1, 2006; 291(1): H327 - H335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. H. Everett IV, E. E. Wilson, S. Foreman, and J. E. Olgin Mechanisms of Ventricular Fibrillation in Canine Models of Congestive Heart Failure and Ischemia Assessed by In Vivo Noncontact Mapping Circulation, September 13, 2005; 112(11): 1532 - 1541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Comtois, J. Kneller, and S. Nattel Of circles and spirals: Bridging the gap between the leading circle and spiral wave concepts of cardiac reentry Europace, January 1, 2005; 7(s2): S10 - S20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Szumowski, P. Sanders, F. Walczak, M. Hocini, P. Jais, R. Kepski, E. Szufladowicz, P. Urbanek, P. Derejko, R. Bodalski, et al. Mapping and ablation of polymorphic ventricular tachycardia after myocardial infarction J. Am. Coll. Cardiol., October 19, 2004; 44(8): 1700 - 1706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. R. Efimov, V. P. Nikolski, and G. Salama Optical Imaging of the Heart Circ. Res., July 9, 2004; 95(1): 21 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Cheng, L. Li, V. Nikolski, D. W. Wallick, and I. R. Efimov Shock-induced arrhythmogenesis is enhanced by 2,3-butanedione monoxime compared with cytochalasin D Am J Physiol Heart Circ Physiol, January 1, 2004; 286(1): H310 - H318. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. G. Akar and D. S. Rosenbaum Transmural Electrophysiological Heterogeneities Underlying Arrhythmogenesis in Heart Failure Circ. Res., October 3, 2003; 93(7): 638 - 645. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Iravanian, Y. Nabutovsky, C.-R. Kong, S. Saha, N. Bursac, and L. Tung Functional reentry in cultured monolayers of neonatal rat cardiac cells Am J Physiol Heart Circ Physiol, June 5, 2003; 285(1): H449 - H456. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. I. Berul Electrophysiological phenotyping in genetically engineered mice Physiol Genomics, May 13, 2003; 13(3): 207 - 216. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-R. Choi, F. Burton, and G. Salama Cytosolic Ca2+ triggers early afterdepolarizations and torsade de pointes in rabbit hearts with type 2 long QT syndrome J. Physiol., September 1, 2002; 543(2): 615 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Boersma, Z. Zetelaki, J. Brugada, and M. Allessie Polymorphic Reentrant Ventricular Tachycardia in the Isolated Rabbit Heart Studied by High-Density Mapping Circulation, June 25, 2002; 105(25): 3053 - 3061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Lee, S.-F. Lin, T. Ohara, C. Omichi, Y. Okuyama, E. Chudin, A. Garfinkel, J. N. Weiss, H. S. Karagueuzian, and P.-S. Chen Effects of diacetyl monoxime and cytochalasin D on ventricular fibrillation in swine right ventricles Am J Physiol Heart Circ Physiol, June 1, 2001; 280(6): H2689 - H2696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. H Samie and J. Jalife Mechanisms underlying ventricular tachycardia and its transition to ventricular fibrillation in the structurally normal heart Cardiovasc Res, May 1, 2001; 50(2): 242 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Patwardhan, S. Moghe, K. Wang, and F. Leonelli Frequency modulation within electrocardiograms during ventricular fibrillation Am J Physiol Heart Circ Physiol, August 1, 2000; 279(2): H825 - H835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. H. Samie, R. Mandapati, R. A. Gray, Y. Watanabe, C. Zuur, J. Beaumont, and J. Jalife A Mechanism of Transition From Ventricular Fibrillation to Tachycardia : Effect of Calcium Channel Blockade on the Dynamics of Rotating Waves Circ. Res., March 31, 2000; 86(6): 684 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. V. Zaitsev, O. Berenfeld, S. F. Mironov, J. Jalife, and A. M. Pertsov Distribution of Excitation Frequencies on the Epicardial and Endocardial Surfaces of Fibrillating Ventricular Wall of the Sheep Heart Circ. Res., March 3, 2000; 86(4): 408 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Chen, R. Mandapati, O. Berenfeld, A. C. Skanes, and J. Jalife High-Frequency Periodic Sources Underlie Ventricular Fibrillation in the Isolated Rabbit Heart Circ. Res., January 7, 2000; 86(1): 86 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Vaidya, G. E. Morley, F. H. Samie, and J. Jalife Reentry and Fibrillation in the Mouse Heart : A Challenge to the Critical Mass Hypothesis Circ. Res., July 23, 1999; 85(2): 174 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Qu, J. N. Weiss, and A. Garfinkel Cardiac electrical restitution properties and stability of reentrant spiral waves: a simulation study Am J Physiol Heart Circ Physiol, January 1, 1999; 276(1): H269 - H283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Mandapati, Y. Asano, W. T. Baxter, R. Gray, J. Davidenko, and J. Jalife Quantification of Effects of Global Ischemia on Dynamics of Ventricular Fibrillation in Isolated Rabbit Heart Circulation, October 20, 1998; 98(16): 1688 - 1696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Athill, T. Ikeda, Y.-H. Kim, T.-J. Wu, M. C. Fishbein, H. S. Karagueuzian, and P.-S. Chen Transmembrane Potential Properties at the Core of Functional Reentrant Wave Fronts in Isolated Canine Right Atria Circulation, October 13, 1998; 98(15): 1556 - 1567. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Skanes, R. Mandapati, O. Berenfeld, J. M. Davidenko, and J. Jalife Spatiotemporal Periodicity During Atrial Fibrillation in the Isolated Sheep Heart Circulation, September 22, 1998; 98(12): 1236 - 1248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lee, G. Morley, Q. Huang, A. Fischer, S. Seiler, J. W. Horner, S. Factor, D. Vaidya, J. Jalife, and G. I. Fishman Conditional lineage ablation to model human diseases PNAS, September 15, 1998; 95(19): 11371 - 11376. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Medkour, A. E. Becker, K. Khalife, and J. Billette Anatomic and Functional Characteristics of a Slow Posterior AV Nodal Pathway : Role in Dual-Pathway Physiology and Reentry Circulation, July 14, 1998; 98(2): 164 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Janse, T. Opthof, and A. G. Kleber Animal models of cardiac arrhythmias Cardiovasc Res, July 1, 1998; 39(1): 165 - 177. [Full Text] [PDF]
|
|
|
|
|
|
O. Berenfeld and J. Jalife Purkinje-Muscle Reentry as a Mechanism of Polymorphic Ventricular Arrhythmias in a 3-Dimensional Model of the Ventricles Circ. Res., June 1, 1998; 82(10): 1063 - 1077. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ikeda, M. Yashima, T. Uchida, D. Hough, M. C. Fishbein, W. J. Mandel, P.-S. Chen, and H. S. Karagueuzian Attachment of Meandering Reentrant Wave Fronts to Anatomic Obstacles in the Atrium : Role of the Obstacle Size Circ. Res., November 19, 1997; 81(5): 753 - 764. [Abstract] [Full Text]
|
|
|
|
|
|
T. Ikeda, L. Czer, A. Trento, C. Hwang, J. J. C. Ong, D. Hough, M. C. Fishbein, W. J. Mandel, H. S. Karagueuzian, and P.-S. Chen \E Induction of Meandering Functional Reentrant Wave Front in Isolated Human Atrial Tissues Circulation, November 4, 1997; 96(9): 3013 - 3020. [Abstract] [Full Text]
|
|
|
|
|
|
R. A. Gray, G. Ayers, and J. Jalife Video Imaging of Atrial Defibrillation in the Sheep Heart Circulation, February 18, 1997; 95(4): 1038 - 1047. [Abstract] [Full Text]
|
|
|
|
|
|
R. A. Gray, A. M. Pertsov, and J. Jalife Incomplete Reentry and Epicardial Breakthrough Patterns During Atrial Fibrillation in the Sheep Heart Circulation, November 15, 1996; 94(10): 2649 - 2661. [Abstract] [Full Text]
|
|
|
|
|
|
D. M. Roden, R. Lazzara, M. Rosen, P. J. Schwartz, J. Towbin, and G. M. Vincent Multiple Mechanisms in the Long-QT Syndrome: Current Knowledge, Gaps, and Future Directions Circulation, October 15, 1996; 94(8): 1996 - 2012. [Abstract] [Full Text]
|
|
|
|
|
|
Y. Cheng, K. A. Mowrey, V. Nikolski, P. J. Tchou, and I. R. Efimov Mechanisms of shock-induced arrhythmogenesis during acute global ischemia Am J Physiol Heart Circ Physiol, June 1, 2002; 282(6): H2141 - H2151. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||